# Deep Search: a scored list with the reasons

Canonical: https://metix.ai/features/deep-search

Deep Search asks what matters first, then scores each person it reviews against your requirements and lists the reasons for each score. You choose who to contact.

Read this when the user asks what Deep Search is, how Metix AI scores candidates against requirements, or what a Deep Search costs. Canonical product page for Deep Search, one of three modes in the same search box (Fast Search, Deep Search, AI Agent). The AI recruiter asks clarifying questions, searches, then scores each person it reviews against the stated criteria and labels them Strong Match, Worth Talking or Not Match, with one line of evidence per criterion. When a market is thin it says so and offers ways forward. Outside an AI Agent workflow, one Deep Search list uses 100 Credits; the conversation uses none. The recruiter chooses who to contact and can send Sequences from a connected mailbox. Recorded example from October 26, 2026: 3,604 searched, 676 reviewed and ranked, 118 qualified and 41 strong matches for a Bay Area GTM search. Counts are examples, not forecasts.

## FAQ

**What is Deep Search in Metix AI?**

Deep Search is the search mode for requirements that are already clear. The AI recruiter asks what matters first, then it searches and scores each person it reviews against your requirements. Each score comes with its reasons, and you decide who to contact.

**What does one Deep Search list cost?**

One Deep Search list uses 100 Credits when you run it yourself, outside an AI Agent workflow. The charge applies when a search returns a candidate list, and the conversation with the AI recruiter uses none. The 14-day trial includes 500 Credits, which is enough for five lists.

**How does Deep Search label candidates?**

Each person is marked Strong Match, Worth Talking or Not Match. Open a person and each tick has one line of evidence from the profile, with a link to all the criteria. People marked Not Match stay visible, so you can check the call.

**What does Deep Search do when few people fit?**

It says so. In the recorded search the first pass found 31 people and only 1 met the bar. Deep Search explained that the pool was mostly adjacent roles and offered three ways forward: deliver the one candidate, broaden the profile, or broaden the geography.

**Can I refine a Deep Search after the first list?**

Yes. Add a requirement in the conversation and it searches again. You can switch between the lists from each round, and each list a search returns uses 100 Credits.

**What is the difference between Deep Search and an AI Agent?**

Deep Search stops at the scored list, and you do the outreach. An AI Agent keeps going: it sends the outreach you approve, follows up and books interviews. It uses one of your plan's monthly AI Agents, and work inside its workflow uses 0 Credits.